Where have all the houses gone ?
Some approaches to Beaker Settlement by Richard Bradley
P OSSIBLY the greatest problem in the archaeology of the Neolithic period is the extreme rarity of recognisable house sites. In this short article I would like to consider the ways in which this problem has affected our knowledge of Beaker settlement in this country. This is not intended as a general survey of Beaker houses simply because so few are yet known that a full analysis of the continental evidence is needed before generalisations can usefully be made. This article is concerned instead with a few of the problems which grow out of the seeming absence of Beaker houses, and is really an oblique commentary upon archaeological methods and preoccupations. The merits of these various approaches will be considered in the light of excavation on a Beaker settlement at Belle Tout in East Sussex. Since a full discussion of this site is already in the press what is said here does not cover the same ground as the final report (P.P.S. 1970).
Risby Warren in East Anglia may be taken as a typical Beaker living site with which to begin this account. This is one of a scatter of small domestic complexes and simply consists of a patch of discoloured sand, a hearth and a group of shallow pits ; no post holes were recorded. The fact that this pattern was so usual at this date at first led to attempts to explain the absence of house sites in theoretical terms. Since no houses were found the Beaker settlers were assigned to tents which were assumed to leave no trace in the ground. In fact shelters of much this type do remain from
Richard Bradley read law at Oxford, where he was President of the OUAS. He is now at Southampton doing research on the archaeological potential of the area of South Hampshire scheduled for London overspill.
Mesolithic contexts and 'tent rings' remain from the Palaeolithic ; but this theory found favour and the Beaker immigrants were relegated to a shadowy role as pastoral nomads. The same factors led on to the suggestion that the absence of houses could make them prospectors or traders. One writer thought they were pirates. A more sophisticated version of this line of reasoning has been the assumption that Beaker groups were engaged in a regular cycle of seasonal movements. This theory may be true of some sites but the evidence from Belle Tout for autumn and spring sown crops seems to rule out this line of argument. Another more satisfactory explanation of the scatter of small nuclei like those about Risby Warren is that continuously occupied domestic complexes regularly shifted site as a result of rotting house footings and contaminated storage pits. This pattern has been described as 'settlement drift'.
If these are strictly theoretical ways of explaining why the Beaker people never had houses, practical arguments have been proposed which allow them houses to live in, but do not allow us to find them. For example the fact that pottery of this date will not normally survive close to the surface may have persuaded us that scatters of worked flint in isolation are only knapping sites. More important it is quite possible for domestic buildings to have been raised upon sill beams, or built upon footings of turf or unfired clay. In no case would these leave much trace in the ground, and so all these arguments have the comfortable virtue that they cannot actually be proved wrong. A more reasonable explanation is that, whether Beaker houses existed or not, the scouring effects of soil erosion would have lowered the surface of the ground to such an extent that post holes would not survive unless the thickness of soil above had sealed them off from the elements. More than 30 centimetres of soil erosion have been detected on some sites, and as a result, a scatter of artefacts and the bottoms of truncated pits are usually all that await the excavator. This is of course so discouraging that only one Beaker domestic site in England has been fully stripped.
This second argument does contain one valuable loophole, for it should be possible to record structural remains on sites where later deposits have sealed the occupation levels from soil erosion. It is for this reason that structural remains can sometimes be recognised under barrows but cannot be traced beyond the limits of the covering mound. The same applies to sites sealed below deep deposits of valley drift. One example is Downton, in Wiltshire, where a number of clear Beaker post holes were identified in the excavation of a Roman villa. Not surprisingly the argument has been extended, and the view advanced that Beaker settlements are normally in valleys, and not upon the hills where we might first look for them. Though this may explain the 'false